Is it just me, or are current LLM benchmarks failing to capture actual usability? (Gemma 4 vs. Gemini/Claude Opus)
英文标题：Is it just me, or are current LLM benchmarks failing to capture actual usability? (Gemma 4 vs. Gemini/Claude Opus)
Disclaimer, this was kinda written with AI (Gemma 4 again) but it also did really well here, it outputted what I wanted, and when I asked it to refine stuff or improve on certain areas it did that without compromising others or making things bulky. I’ve been noticing a massive disconnect lately betw
打开原文相关 3 · 机会 0
中文
Disclaimer, this was kinda written with AI (Gemma 4 again) but it also did really well here, it outputted what I wanted, and when I asked it to refine stuff or improve on certain areas it did that without compromising others or making things bulky. I’ve been noticing a massive disconnect lately betw
中文全文翻译正在补齐，可先查看下方英文原文。
English / 原文
Disclaimer, this was kinda written with AI (Gemma 4 again) but it also did really well here, it outputted what I wanted, and when I asked it to refine stuff or improve on certain areas it did that without compromising others or making things bulky. I’ve been noticing a massive disconnect lately betw